Readiness of US General Surgery Residents for Independent Practice

Brian C. George, Jordan D. Bohnen, Reed G. Williams, Shari L. Meyerson, Mary C. Schuller, Michael J. Clark, Andreas H. Meier, Laura Torbeck, Samuel P. Mandell, John T. Mullen, Douglas S. Smink, Rebecca E. Scully, Jeffrey G. Chipman, Edward D. Auyang, Kyla P. Terhune, Paul E. Wise, Jennifer N. Choi, Eugene F. Foley, Justin B. Dimick, Michael A. ChotiNathaniel J. Soper, Keith D. Lillemoe, Joseph B. Zwischenberger, Gary L. Dunnington, Debra A. Darosa, Jonathan P. Fryer

Research output: Contribution to journalArticlepeer-review

240 Scopus citations

Abstract

Objective: This study evaluates the current state of the General Surgery (GS) residency training model by investigating resident operative performance and autonomy. Background: The American Board of Surgery has designated 132 procedures as being "Core" to the practice of GS. GS residents are expected to be able to safely and independently perform those procedures by the time they graduate. There is growing concern that not all residents achieve that standard. Lack of operative autonomy may play a role. Methods: Attendings in 14 General Surgery programs were trained to use a) the 5-level System for Improving and Measuring Procedural Learning (SIMPL) Performance scale to assess resident readiness for independent practice and b) the 4-level Zwisch scale to assess the level of guidance (ie, autonomy) they provided to residents during specific procedures. Ratings were collected immediately after cases that involved a categorical GS resident. Data were analyzed using descriptive statistics and supplemented with Bayesian ordinal model-based estimation. Results: A total of 444 attending surgeons rated 536 categorical residents after 10,130 procedures. Performance: from the first to the last year of training, the proportion of Performance ratings for Core procedures (n = 6931) at "Practice Ready" or above increased from 12.3% to 77.1%. The predicted probability that a typical trainee would be rated as Competent after performing an average Core procedure on an average complexity patient during the last week of residency training is 90.5% (95% CI: 85.7%-94%). This falls to 84.6% for more complex patients and to less than 80% for more difficult Core procedures. Autonomy: for all procedures, the proportion of Zwisch ratings indicating meaningful autonomy ("Passive Help" or "Supervision Only") increased from 15.1% to 65.7% from the first to the last year of training. For the Core procedures performed by residents in their final 6 months of training (cholecystectomy, inguinal/femoral hernia repair, appendectomy, ventral hernia repair, and partial colectomy), the proportion of Zwisch ratings (n = 357) indicating near-independence ("Supervision Only") was 33.3%. Conclusions: US General Surgery residents are not universally ready to independently perform Core procedures by the time they complete residency training. Progressive resident autonomy is also limited. It is unknown if the amount of autonomy residents do achieve is sufficient to ensure readiness for the entire spectrum of independent practice.

Original languageEnglish (US)
Pages (from-to)582-594
Number of pages13
JournalAnnals of surgery
Volume266
Issue number4
DOIs
StatePublished - Oct 1 2017

Bibliographical note

Funding Information:
This study was funded by a grant from the American Board of Surgery. The initial development of SIMPL was funded via grants from Massachusetts General Hospital, Northwestern University, and Indiana University. Later development was funded by contributions from the members of the Procedural Learning and Safety Collaborative (PLSC, http://www.procedurallearning.org).

Publisher Copyright:
© 2017 Wolters Kluwer Health, Inc. All rights reserved.

Keywords

  • assessment
  • autonomy
  • independent practice
  • performance
  • readiness
  • resident
  • simpl
  • surgical education

Fingerprint

Dive into the research topics of 'Readiness of US General Surgery Residents for Independent Practice'. Together they form a unique fingerprint.

Cite this